Ur5
PulseAugur coverage of Ur5 — every cluster mentioning Ur5 across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
TCP4D model unifies robot manipulation tracking and generation
Researchers have developed TCP4D, a novel model designed for robot manipulation that unifies tracking and generation of tool-center-point (TCP) poses from monocular RGB video. This model can recover metric TCP poses in …
-
MITO dataset and simulator advance non-line-of-sight perception with mmWave sensing
Researchers have introduced MITO, a novel dataset and simulator designed to enhance non-line-of-sight (NLOS) perception using millimeter-wave (mmWave) sensing. This dataset comprises over 24 million mmWave frames and 55…
-
Google DeepMind unveils Genie 4 world model with dynamic environment generation
Google DeepMind has released Genie 4, a new world model capable of generating interactive environments and integrating with Gemini models for task execution. Genie 4 allows for dynamic updates to the environment based o…
-
RoboTracer system enhances robot trajectory tracking with 3D spatial reasoning
Researchers have developed RoboTracer, a novel system that enhances robot trajectory tracking by leveraging 3D spatial perception and metric reasoning. Unlike previous methods that directly output joint torques or rely …
-
New SA-WAM model integrates 3D data into robot policy learning
Researchers have developed a Spatially Aware World Action Model (SA-WAM) that integrates 3D geometric information into large-scale pretrained video diffusion models for robot policy learning. This model repurposes exist…
-
Embodied AI shifts to "imagination" with world models at RSS 2026 · 1 source tracked
At RSS 2026, the focus for embodied AI shifted from data-driven imitation to "imagination" powered by world models. Researchers are moving beyond simply generating realistic videos to using world models as a "logical co…
-
Robotic grasping improved with deep reinforcement learning
Researchers have developed a novel reinforcement learning framework to improve robotic grasping capabilities. This system integrates a Deep Q-Network (DQN) with keypoint-based object representations, using 2D images to …